A fuzzy portfolio trading system in the foreign exchange market based on technical analysis

Technical trading rules are extensively used by foreign exchange (forex) traders. Despite the essential need to the forex diversification, it is not addressed by academic researches to generate forex portfolio trading systems based on technical indices. This paper aims to develop an interpretable and accurate Takagi-Sugeno-Kang (TSK) system for forex portfolio trading. The system uses technical indices of the forex rates and delivers the preferred portfolio composition among multiple foreign currencies. The proposed model considers the transaction cost and trading risk, which are the two important factors in the high frequency trading strategies. The proposed model was implemented to develop a trading system for portfolio trading among the five of the most traded currencies in the Tehran forex market. Four experiments were designed to examine the performance of the proposed model in different market trends, in terms of the portfolio return and risk adjusted return. According to the experimental results, the proposed model is able to extract profitable portfolio trading systems in this market, especially when the market is in the downward trend.